Bladder Cancer Immunotherapy: Progress and Current Limitations Donald L. Lamm, MD, FACS Bladder Cancer, Genitourinary Oncology Phoenix, AZ Hebrew University

Bladder Cancer Immunotherapy: Progress and Current Limitations

Donald L. Lamm, MD, FACS

Bladder Cancer, Genitourinary Oncology

Phoenix, AZ

Hebrew University of Jerusalem

Yissum Technology Transfer

Tel Aviv, September 27, 2005

www.BCGOncology.com

Bladder Cancer Statistics, 2005

• New Cases: 63,210 Men: 47,010; #4 Women: 16,200 #8

• Estimated Deaths: 13,180 Men: 8,970; #9 Women: 4,210

• Incidence/Mortality: 20.8% Men: 19% Women: 26%

• Prevalence: More than 600,000 in US

Bladder Cancer, 2005

• Peak Onset: 6th to 8th decades

• Men/women: 3 to 1

• Twice as common in white men compared with African American men

• Genetic mutations: genes on chromosome 9 including p16. Invasion p53, Rb, p21. H19: 84%

• Screening: hematuria detection reduces mortality

Diagnosis

• 85% present with gross of microscopic hematuria

• Cystoscopy is key: papillary tumors are easily seen. High grade, solid, flat or in situ tumors may not be seen

• Urinary Cytology: 80% + sensitivity in high grade tumors with 95% specificity. Insensitive with low grade. Sensitivity improved with FISH

• IVP, CT scan for upper tract evaluation

Cystoscopy showing bladder tumor

TURBT

Bladder Cancer Immunotherapy is Primarily BCG Immunotherapy

Goals• Brief History of BCG

• BCG Controlled Trials: vs TUR alone, vs Chemo

• Improving BCG Therapy: Maintenance, BCG + Ifn

• Limitations of BCG

• Prospects for New Agents

BCG History

• 1921- Calmette & Guerin successfully tame M. bovis

• 1929- Pearl reports TB reduces incidence of CA

• 1930- Lubeck incident brings erroneous scandal

• 1935- Holmgren reports BCG success in 28 cancer pts

• 1936- Rosenthal reports BCG’s profound RE stimulation

• 1950’s- Animal studies confirm efficacy

• 1972- Rosenthal reports leukemia with vaccination

• 1970’s- Multiple uncontrolled reports of clinical efficacy

BCG History- Bladder Cancer

• 1976- Morales reports 12 fold reduction in recurrence in 9 patients treated with BCG

• 1973- Lamm begins controlled animal studies in TCC

• 1978-NCI controlled trials begin based on Morales’ work

• 1980- Lamm reports first successful controlled trial

• 1982- Current: Brosman, Netto, Martinez-Pineiro and many others report BCG to be superior to Chemotherapy

Lamm, DL:Invest Urology 14:369, 1977

Lamm, DL: J Urol 124(1):38-40, 1980

Lamm, DL: J Urol 134(1):40-47, 1985.

Lamm DL: N Engl J Med. 1991;325:1205

Per

cen

tag

e o

f p

atie

nts

0 12 24 36 48 60 72Time after registration, months

100

80

60

40

20

0

n 5-year RFS

BCG CIS 64 45% BCG Ta, T1 63 37% Doxorubicin Ta, T1 67 18%Doxorubicin CIS 68 17%

BCG Versus Doxorubicin: Time to Treatment Failure

Diet, Lifestyle and Environmental Factors

• Diet: low vitamin A, low serum carotene increase risk; increased fat increases risk; soy, garlic, selenium, NSAIDS, and green tea may reduce risk

• Vitamins may be protective: A (differentiating agent); B6; C (antioxidant); E (antioxidant), and possibly folic acid and D

Kaplan Meier Estimate of 5 Year Tumor Free Rate

Lamm D. J Urol 151(1): 21-26, 1994100

90

80

70

60

50

40

30

20

10

0

Months After Registration0 5 10 15 20 25 30 35 40 45 50 55 60

In 65 Patients Receiving Vitamin Supplement and BCG TherapyFor Bladder Carcinoma

Per

cent

Tum

or F

ree

40,000u Vitamin A, 100mg B6, 2gm C, 400mg E: "Oncovite"

p=0.0014

RDA VitaminsRDAVitaminsOncovite

(N=30)(N=35)

Oncovite (Vitamins A, B6, C &E) in Bladder Cancer

• Overall recurrence reduced from 80% to 40% (P=0.0011)

• 42% reduction in recurrence in Ta, T1 TCC

• 53% reduction in low grade (G1, G2) TCC

• Associated with statistically significant increase in long-term NK cell activity in BCG treated patients

Controlled BCG Trials

Author no. NoRx BCG Ben. PLamm '85 57 52% 20% 32% <.001Herr '85 86 95% 42% 53% <.001Herr (CIS) '86 49 100% 35% 65% <.001Yamamoto'90 44 67% 17% 50% <0.05Pagano '91 133 83% 26% 57% <.001Mekelos '93 94 59% 32% 27% <0.02Krege'96 224 48% 29% 24% <0.05Kolodziej ‘02 155 55% 19% 36% <.001Total: 842 70% 27% 43%

Meta-Analysis of BCG vs. TUR AloneShelly et al. Cochrane Group BJU Int 2001, 88:209

• 26 publications reviewed

• 6 acceptable trials with 585 patients

• Mean log hazard ratio for recurrence -.83, P<0.001

• 56% reduction in hazard attributable to BCG

• Manageable toxicity: cystitis 67%, hematuria 23%, fever 25%, frequency 71%

• Conclusion: BCG provides significantly better prophylaxis of tumor recurrence in Ta, T1 TCC

Randomized BCG vs. Chemotherapy Studies

BCG0

7%13%

Rec Adv. P valueThiotepa

Authorvsvsvs

Chemo47%43%36%

+47+35+26

<.01<.01<0.05

Brosman '82Netto '83Martinez '90

Doxorubicin53%13%24%

vsvsvs

78%43%42%

+21+30+18

<.02<.01<.05

Lamm '91Martinez '90Tanaka '94

Epirubicin33% vs 47% +14 <.0001 vd Meijden '01

Randomized BCG vs. MMC StudiesBCG

42861476446435124383213

Rec. BCG P valuevsvsvsvsvsvsvsvsvsvsvsvs

MMC+30+34+19-5-21-3+9+15+5+24+22+13

<.01<.001

NSNS

NS<.01<.01

NS<.001<.001<.01

%%%%%%%%%%%%

36.7% of 781 vs 53.8% of 771 (+17%) in maintenance BCG studies. 6/6 maintenance BCG studies significant vs 1/5 non-maint.

346280424243566629625426

%%%%%%%%%%%%

Author/yearPagano '87Finnblad '89Lee '92Witjes '94Vegt '95 '95SWOG '96Malmstr. '96Krege '96Ayed '98Milan '00Nogueira '01

60055-23-N

BCG Versus Mitomycin-C (SWOG 8795)

Time To Recurrence

Per

cent

Rec

urre

nce

363024181260

100

90

80

70

60

50

40

30

20

10

0

BCGMMC

190187

4464

Not Reached20

At. Risk FailMedian

in Months

Lamm DL: Urol Oncol 1:119-126, 1995

Intravesical BCG is superior to mitomycin C in reducing tumour recurrence in high-risk superficial bladder cancer:

a meta-analysis of randomized trials. Shelley et al. (2004) BJU Int. 93:485-90

“This is the highest level of evidence-based medicine and the results presented here suggest that intravesical BCG is superior to mitomcycin C.”

“A subgroup analysis of 3 trials that included only high-risk Ta and T1 patients indicated no heterogeneity (P-0.25) and a LHR for recurrence of -0.371 (0.012). With MMC used as the control in the meta-analysis, a negative ratio is in favour of BCG and, in this case, was highly significant (P<0.001).”

Complete Response in CISIntravesical Chemotherapy

Agent N CR% Range

Thiotepa 89 38% (20-50%)

Adriamycin 212 48% (0-88%)

Mitomycin C 196 53% (0-100%)

Epirubicin 84 56%

Epi + MMC 21 81%

REFERENCE IN SITU INVASION (%) YEARS

Melamed et al (1964) 25 9 (36%) <5

Koss et al. (1979) 13 7 (54%) 1 to 6

Kulatilake et al (1970) 5 3 (60%) 2

Utz et al (1970) 62 37 (60%) <5

Farrow, et al (1977) 58 8 (14%) <5

Sharma et al (1970) 17 14 (82%) NA

Yates-Bell (1971) 3 3 (100%) <3

Barlebo et al (1972) 10 0 (0%) NA

Anderson et al (1973) 15 12 (80%) NA

Skinner et al (1974) 59 49 (83%) NA

Riddle et al (1974) (diffuse)

23 18 (78%) 1 to 11

(focal) 13 1 (8%) 1 to 16

Althausen et al (1976) 12 10 (83%) 1.5

Starklint et al (1976) 43 23 (53%) >1

Herr (1983) 24 23 (50%) 1 to 3

Total 382 206 (54%) --

Progression in CIS Prior to BCG Immunotherapy

Comparison of BCG Preparations inthe Treatment of CIS

ConnaughtTokyoPasteurTiceEvansA FrappierS AfricanDanishRomanianRIVM

BCG Prep

45011123027718014513424215

N

79%77%74%71%65%60%69%67%64%60%

CR %

70% - 92%63% - 84%40% - 80%56% - 88%53% - 88%39% - 100%

Total:1496 (72%)39% - 100%

Range CR

BCG vs Chemo For CIS: Meta-AnalysisSylvester: J Urol. 174:86, 2005

• 9 randomized trials including 700 pts. With CIS

• Chemo: MMC, Epi, Adria, or sequential MMC/Adria

• BCG: 68% CR vs Chemo CR: 52%; P=0.0002

• 3.6 year follow: 47% BCG vs 26% Chemo NED

• 26% reduction in disease progression with BCG

• “BCG reduces the risk of short and long-term treatment failure compared with chemotherapy… agent of choice in the treatment of CIS.”

Meta-analysis of BCG versus Chemotherapy in CIS

Sylvester RJ: J Urol. 2005 Jul;174(1):86-91

Carcinoma in situ: SWOG 8507

CIS: 269 Randomized

114 Induction - 230 Evaluable - 116 Maintenance

6 week BCG 6 week BCG

3 mo: 58% CR P=0.7 55% CR Observation 3 week BCG

6 mo: 69% CR P=0.01 84% CR

26% of CIS failures at 3 mo NED at 6 without further treatment; 64% with 3wk BCG

3 Week Maintenance BCG in 550 Randomized, 385 Evaluable Patients

Recurrence -freeSurvival Survival

Worsening -freeSurvival

Lamm DL et al, J Urol 163, 1124, 2000

p < 0.0001 p = 0.08p = 0.04

BCG Maintenance: Not Created Equal

YearsCompletion of Therapy*Apparent Increase in Rate of Recurrence One Year After Completion of Maintenance

**

Per

cent

Tum

or R

ecur

renc

e

100

90

80

70

60

50

40

30

20

10

0

M, TaT1, 3wk maintenance BCG M, CIS, 3wk maintenance BCG I, CIS, 6wk induction BCG I, TaT1, 6wk induction BCG

M. Ta, T1

M. CIS

I. CIS

I. Ta, T1

0 1 2 3 4 5 6 7 8 9* **

N=385, 3q 3-6mo.

Time in months

Glo

bal r

ecur

renc

e

Maintenance

0 12 24 36 48 60 72

1.0.9.8.7.6.5.4.3.2.1

0.0 Control

N=126, 6q 6mo.

Months

90100

7080

60

4050

3020

010

0 369 18 27

% D

isea

se F

ree

N=93 pts. 1q 1mo.

M BCGI BCG

Months

100

50

0

% T

umor

Fre

e

3 2112 15 18 3324 27 306 9

M BCGI BCG

N=42 pts. 1q 3mo.

3 Weekly Maintenance BCG Schedule: Lamm 2005

Induction Mo: 3 6 12 18 24 36 Yr: 4 5 6 8 10 12

Full x6 1/3x3: * * * * * * * * * * * *

Full strength BCG is given weekly for 6 weeks during induction (reduced if needed for increased side effects)

1/3 BCG, reduced to 1/10, 1/30, 1/100th if needed due to increased side effects, given at 3,6,12,18,24, and 36 months, then years 4, 5, 6, 8, 10 and 12 years in G3/CIS

Dose-Response Curve to BCG (in mice)

Individual responses and preparations vary, buttoo little or too much BCG reduces effect

Lamm DL, et al. J Urol. 1982; 128: 1104-1108.

105 106 107 108

60

40

20

0

-20

BCG ·colony forming units

Incr

ease

d su

rviv

al v

s co

ntro

l % PasteurTiceGlaxoOver all

Progression: Maintenance BCG

Patients No BCG BCG ORNo Maint 1049 10.3% 10.8% 1.28Maintenance 3814 14.7% 9.5% 0.63

Test for heterogeneity: P = 0.008

BCG was only effective in trials with maintenance, where it reduced the risk of progression by 37%

p = 0.00004.

Sylvester RJ: J Urol. 2002 Nov;168(5):1964-70.Meta Analysis of 24 Randomized Trials

Study Publ YearAuthor and Group

Events / PatientsNo BCG BCG

Statistics (O-E) Var.

OR & CI:(BCG No BCG)

|1-OR|% ± SD

ProgressionAll Studies With Maintenance

1991 Pagano (Padova) 11 / 63 3 / 70 -4.4 3.1


1987 Badalament (MSKCC) 6 / 46 6 / 47 -0.1 2.6


2000 Lamm (SW8507) 102 / 192 87 / 192 -7.5 24.1


2001 Palou 2 / 61 3 / 65 0.4 1.2


1996 Rintala (Finnbl 2) 3 / 90 3 / 92 0 1.5


1995 Rintala (Finnbl 2) 4 / 40 2 / 28 -0.5 1.3


1995 Lamm (SW8795) 24 / 186 15 / 191 -4.8 8.8


1999 Malmstrom (Sw-N) 22 / 125 15 / 125 -3.5 7.9


2001 Nogueira (CUETO) 8 / 127 10 / 247 -1.9 3.9


1991 Rintala (Finnbl 1) 2 / 58 3 / 51 0.7 1.2


2001 de Reijke (EORTC) 18 / 84 10 / 84 -4 5.9


2001 vd Meijden (EORTC) 19 / 279 24 / 558 -4.7 9.1


1982 Brosman (UCLA) 0 / 22 0 / 27 0 0


1990 Martinez-Pineiro 4 / 109 1 / 67 -0.9 1.2


1999 Witjes (Eur Bropir) 2 / 25 1 / 28 -0.6 0.7


1997 Jimenez-Cruz 7 / 61 6 / 61 -0.5 2.9


1994 Kalbe 2 / 35 0 / 32 -1 0.5


1991 Kalbe 2 / 17 0 / 21 -1.1 0.5


1993 Melekos (Patras) 7 / 99 2 / 62 -1.5 2


1988 Ibrahiem (Egypt) 12 / 30 5 / 17 -1.1 2.6

Total 257 / 1749 196 / 2065 -36.8 80.9(14.7 %) (9.5 %)

37% ±9reduction

0.0 0.5 1.0 1.5 2.0BCG No BCGTest for heterogeneitybetter better2

=9.73, df=18: p=0.9Treatment effect: p=0.00004

Survival

Death Patients No BCG BCG Total ORAll 2930 26.7% 23.2% 24.8% 0.89Bladder 2370 7.7% 5.6% 6.5% 0.81

The reductions in the odds of death, 11% overall and 19% bladder cancer, are not statistically significant, as might be expected with 2.5 year mean follow up

Sylvester RJ: J Urol. 2002 Nov;168(5):1964-70. Meta Analysis of 24 Randomized Trials

Limitations of BCG Immunotherapy: 50 to 80% Eventually Fail

• Early failure to respond Excess or remote tumors Rapidly dividing/growing tumor Low grade, “non-antigenic” tumors Unresponsive host

• Late recurrence: immunosuppression, resistance

• Toxicity

Treatment of BCG Failure

• Chemotherapy for BCG failures provides poor response rates

19% for MMC post BCG Malmstrom, J Urol, 2001

• Low Dose BCG after one cycle BCG failure provides 60% durable CR (same as BCG naive)

Maymi et al, AUA Abstract 918

Esuvaranathan: Singapore Randomized Trial- Full Dose BCG v 1/3 BCG v 1/3 BCG+Ifn alpha

65 patients randomized to full dose Evan’s BCG vs. 1/3 dose vs. 1/3 dose BCG plus 10 MU Intron A

9 mo. Rec 20 mo. Rec

Full Dose BCG: 32% 48%

1/3 Dose BCG: 12% 24%

1/3 BCG+ Ifn: 12% 12%*

Subsequent randomization to full dose BCG vs. BCG+Ifn, 130 pts:

Mean TTR 5yr KM% Rec. Free

Full Dose BCG (N=60): 58.5 mo. 51% (49% rec)

1/3 Dose BCG (N=29): 61.8 mo. 66% (34%)

1/3 BCG+ Ifn (N=41): 71.8 mo. 79% (21%)

*P=0.035 vsFull dose BCG

Complications of BCG Therapy in 2,569 Patients

Total Tice Connaught

Fever 75(2.9%) 4.7% 4.7%

G. Prost 23(0.9%) 1.8% 1.0%

Pneum/hep. 18(0.7%) .4% .8%

Arthralgia 12(0.6%) .7% .1%

Hematuria 24(1.0%) .3% .6%

Rash 8(0.3%) .4% 0

Uret. Obstr. 8(0.3%) .6% .4%

Epididymitis 10(0.4%) .4% 0

Contr. blad. 6(0.2%) 0 .3%

Renal abscess 2(0.1%) 0 0

Sepsis 10(0.4%) .1% .4%

Lamm DL. Urol Clin North Am. 1992; 19:565-7

Early Comparison KLH Trials

Treatment R/100 pt mo N Rec %

MMC 9.3 23 39%

KLH 10mg 3.3 21 14%

Epodyl 4.8 46 35%

KLH 20 mg 6.5 38 21%

Jurincic, 1988; Flamm, 1990

Purified vs Crude KLH vs BCG

Treatment Incidence Volume Survival

Pure KLH 4/10 1900mm 5

Crude KLH 0/10** 230 ** 10 **

BCG 2/10* 71 ** 9 *

Saline 8/10 3400 3

* P<0.012; ** P<0.002

Complete Response to KLH by Disease Category

Stage CR (N) CR (%)

CIS 9 50%

Ta, T1, CIS 4 33%

Ta, T1 3 20%

Total 16 36%

Conclusions• Bladder Cancer is immunoresponsive and an

excellent model for drug development.

• BCG immunotherapy is superior to chemotherapy and reduces progression, but 50-80% fail.

• Maintenance schedules, vitamins, and interferon may improve response.

• New agents such as KLH and others hold promise for reduced toxicity, and new approaches such as DNA-based therapy are greatly needed!

H19 Expression in Bladder Cancer

•84% of TCC express H19•Levels are nearly undetectable in surrounding normal urothelium

CIS H19 ISH Color Intensity

G2 TCC with H19 Stain

What is the Best InductionSchedule ?

Six weekly instillations: excellent but clearlysuboptimal

With retreatment, stimulation peaks at 3wks

Immune stimulation peaks at 6 weeks

Continued treatment beyond 6 weeks cansuppress the immune response

Controlled trial of "6" vs "6+3" in CIS shows< CR from 69% to 84% (P<0.01)

Why Not Give MonthlyBCG Maintenance ?

Historical and controlled studies show noadvantage over 6 week induction

Immune suppression may occur

Toxicity is increased over induction

There is no biological or immunologicalrationale for the monthly schedule

Percutaneous BCG ?

Two studies failed to demonstrate benefit

17/55 (31%) recurrence with PPD conversion, 51/82(62%) recurrence with no conversion P=0.0225

40-60% of patients convert PPD skin test afterintravesical BCG

More than 90% convert with I.D. BCG

CR in CIS increased from 49% to 77% with PPDconversion (SWOG, P<0.001)

Lamm '85 and Torrence '88:

Optimal BCG Retreatment

"6+6" should be avoided, unless the intervalsince last treatment has been long (many years)and little or no side effects occurred

For repeat BCG, think "3 plus 3"

If a second six week course is given one cannotdistinguish decreased sensitivity to BCG fromiatrogenic immunosuppression

Toxicity of Maintenance BCG

Log dose reductions (1/3, 1/10, 1/30, 1/100th) or stopping maintenance BCG appears to prevent toxicity

Side effects are not required to receive thebenefit of maintenance BCG

Treatment of BCG Sepsis

Isoniazid 300mg, rifampin 600mg, and ethambutol 1200mg daily plus a fluoroquinolone or an aminoglycoside

Prednisone 40mg daily (higher doses sometimes are required) Taper steroid slowly when patient improves Resume steroids if symptoms recur after taper Continue triple antibiotics for 3-6 months No more BCG

Documents

Bladder Cancer Immunotherapy: Progress and Current Limitations Donald L. Lamm, MD, FACS Bladder Cancer, Genitourinary Oncology Phoenix, AZ Hebrew University